Dealing with Dynamic Systems:
31
aspect. (4) Implicit assumptions: Consider, for example, a variable that does
not change over time. The weight for that relation is assumed to be equal to
one. This kind of knowledge a subject often does not find worth mentioning.
(5) Single vs. multiple models: It is not clear whether a subject follows only
one single model at a given point in time or whether there exist several
models concurrently.
For problems (3) to (5), no solution can be given at present. Problems (1)
and (2), however, can be solved using a quantification of the following kind:
For each causal specification of a subject one first counts whether it belongs
to one of the three classes of knowledge (relational, sign, or numerical) and
whether it is correct or false. Then, for each of the three levels one can
determine the “quality of system identification” (QSI) in terms of the differ-
ence between “hits” (HI) and “false alarms” (FA), weighted by a “guessing”
probability (p) according to the following scheme which closely resembles
the discrimination index Pr from the two-high threshold model for recogni-
tion memory (see Snodgrass & Corwin, 1988; the proposed “correction for
guessing” dates back to Woodworth, 1938):
HI FA
QI = (l-p)---- p--, —p ≤ QI < (l-p) (5)
max(HI) max(FA)
The guessing probability for numerical parameters in a dynamic system
could, for instance, be set to zero. In this case all hits are counted relative to
the maximal number of hits, max(HI). If one sets the guessing probability to
0.5 in the case of sign knowledge, then errors lead to a reduction in the QSI
index for that level. The index for structural knowledge, which serves as a
dependent variable in the following experiments, is called “QSI” (“Quality
of Identification”—high QSI revealing a good score because of high corre-
spondence between implemented and assumed causal relations) and results
from an additive combination of the QSI-values for all three knowledge
levels. A study by Miiller (in press) demonstrates considerable reliability and,
thus, sufficient psychometric quality of this index.
3. Experimental Studies on System Properties
In the following section five experiments on the role of different system
properties serve to illustrate the approach just outlined. The focus of the
experiments is on the role of active intervention into a system vs. pure
observation (Exp. 1), on the influence of different degrees of Eigendynamik
(Exp. 2) and side effects (Exp. 3), and on the effects of presentation mode,
prior knowledge, controllability of the system, and degree of control required
(Exp. 4 & 5). For each of the experiments the presentation includes a
description of the independent as well as dependent variables, subjects,